1. Field of the Invention
The present invention relates to a routing system and method, and more particularly to a routing system and method capable of designating an optimum route by exchanging network route information held by each node with adjacent nodes thereof.
2. Description of the Related Art
Conventionally, in a PNNI (Private Network-Network Interface) which is specified by an ATM Forum, an optimum route from a source node to a destination node is designated by periodically or non-periodically exchanging network route information held by each node with adjacent nodes so as to perform a call control.
A PNNI node (as a local node) periodically transmits resource information to the adjacent nodes thereof. The resource information includes an address of the local node, an address of a terminal accommodated in the local node, available cell rates of transmission lines and the like. Each of the adjacent nodes, which have received the resource information, holds the resource information to make a database by which a status of a PNNI network can be understood. The adjacent nodes further transmit the received information to adjacent nodes thereof. Thus, resource information about all nodes of the PNNI network can be held by each node.
FIG. 1 shows resource information which is held in a PNNI network consisting of nodes A, B and C. The node A holds resource information A1, A2 and A3. The information A1 is node A network prefix information. The information A2 is about a status such as an available cell rate and the like of a line from the node A to the node B. The information A3 is about a status such as an available cell rate and the like of a line from the node A to the node C. The nodes B and C have similar resource information. The resource information held by the nodes A, B and C is exchanged between the nodes A, B and C so that information about the entire network can be held by each node. A detailed description thereof will be given later.
According to a database made based on the above-mentioned resource information, a source/originally-transmitting node in the PNNI network can acknowledge which routes may get to a destination node and select an advantageous route of these routes from a given viewpoint of, for example, cost, cell transfer delay, cell delay variation or the like. This is referred to as a source routing.
FIG. 2 shows an example of the source routing by the PNNI.
FIG. 2 shows three routes (i), (ii) and (iii) which may be used to perform a call setup from a terminal T1 accommodated in a node A to a terminal T3 accommodated in a node C. If the node A is preset to select an advantageous route according to the cell delay, then the node A examines the cell delay item in a routing database DB and designates the route (i) because the cell delay of the route (i) is shortest. Further, when a status of the PNNI network is varied, the PNNI node can detect the variation and issue a notification of the variation in the PNNI network.
FIG. 3 shows a notification of resource information being transmitted in a PNNI network. In the diagram, when resource information of a node A is varied, the node A transmits the varied resource information (namely the resource information A1, A2 and A3) to the other nodes B and C. The nodes B and C further transmit the resource information A1 to A3 received from the node A to their adjacent nodes, namely the node C and the node B, and thus the resource information about the node A can be spread on the whole PNNI network.
Further, a node of the PNNI network is provided with a crankback procedure. When a node (as a routing source node) has missed receiving a notification of a variation in the network status and then performs a routing based on outdated network information, a PNNI node which detects the variation notifies the routing source node so as to perform the crankback procedure. If the crankback procedure is not used, then a call loss may occur and a terminal may need to send a call setup message again so as to reestablish the connection. If the crankback procedure is used, then the terminal can ensure the call through another available route without having to reestablish the connection.
FIG. 4 shows an example of the crankback procedure of the PNNI. The node A sends a setup message to the node C ({circle around (1)}), the node C fails in bandwidth hunt on a line connecting the node A and the node C due to receiving the setup message ({circle around (2)}), the node C sends a release message (REL) including a crankback information element to the node A ({circle around (3)}), and the node A designates an alternate route via the node B by analyzing the received crankback information element ({circle around (4)}). In addition, the crankback information element includes call reject causes (for example, user cell rate not being available, failure in VPCI/VCI hunt, etc), and a call reject occurrence position (reject-occurred node identifier which is equivalent to a network prefix, reject-occurred port of the node and the like).
FIG. 5 shows representative items indicating a resource status which items should be transmitted. These items are regulated by a conventional ATM Forum Private Network-Network Interface Specification Ver 1.0 (hereinafter referred to PNNI) and are formed to be a packet, referred to as PTSE, (PNNI Topology State Element) so as to be delivered as difference information to neighboring nodes. A packet for notifying a line resource status corresponds to A2, A3, B2, B3, C2, and C3 of FIG. 1. A packet for notifying a local node ID corresponds to A1, B1, C1 of FIG. 1. A packet (not shown) for transmitting an address of a locally accommodated terminal is used in a case in which it is needed to clearly notify other nodes of the address of the terminal which is accommodated in the local node. These packets are integrated with respect to the whole network so that a PNNI node can obtain information about routes to another PNNI node (the number of the routes, natures of the respective routes, and the like).
A PNNI node (source node), which receives a calling request from a terminal accommodated therein, designates an optimum route getting to a destination node and produces a list including ID information of transit nodes on the optimum route, a node-passing-through order and information which indicates outgoing link at each node. The list is referred to as xe2x80x9ca route designation listxe2x80x9d.
The source node sends a call setup message including the route designation list to the next node. The next node which has received the call setup message examines the list included in the setup message so as to select an outgoing link through which to forward the setup message.
The calling request may be rejected on the way in a case in which the network status is varied after the setup message is sent, or a notification of a variation in the network status does not yet reach the source node at the time the setup message is sent. In this case, by using the above-mentioned crankback procedure, the call setup via another route is performed again in the network.
FIG. 6 is a block diagram showing an example of a conventional node. In the diagram, a call control unit 17 issues xe2x80x9ca route look-up requestxe2x80x9d to a routing unit 14 when a call setup is requested, and as a response the routing unit 14 notifies the call control unit 17 of a route to the destination. The routing unit 14 collects route information from a route information management unit 12 so as to generate the destination route notification for the call control unit 17. The route information management unit 12 receives route information (resource information) delivered from other nodes, sends local node information and information about the other nodes, and stores these information as the routing information. A route information transmitting/receiving unit 11 delivers information between the other nodes and the route information management unit 12. A route information display unit 15 displays the routing information held by the route information management unit 12 to users.
Conventionally, each node, when resource information therein is varied, notifies other nodes of the variation. The other nodes, which have received the variation information, update their databases according to the received information. Thus, the database only indicates a network resource status obtained at the time when the database is referred to. Hence, as a result of the routing, the call may be rejected due to reasons that cannot be determined from the database, such as non-arrival of resource variation information, receiving incorrect resource information obtained by a node-processing error and the like.
In order to designate an optimum route to a destination node in a network, a source node refers to a database thereof from the given viewpoints of cost, cell transfer delay, cell delay variation and the like which are obtained at the time when the database is referred to. For this reason, when node congestion and line failures frequently occur on a route to be designated, avoiding designating the route cannot be realized. That is, the route is designated at the time when the database is referred to, hence problems that have occurred in the past on the route are taken into consideration in the routing.
Thus, when a call setup is requested for a congested node, the congestion of the node further increases and delay of the call setup further extends due to passing through the congested node. Moreover, an established connection may be abnormally disconnected due to designating a route on which the line failures frequently occur. Furthermore, when a node which advertises incorrect rooting information caused from functional abnormality of itself exists in a network, a source node cannot detect its being abnormal and executes a wrong route selection based on incorrect database information.
It is a general object of the present invention to provide a routing system, in which the above problems are eliminated.
A more specific object of the present invention is to provide a routing system and method capable of designating an optimum route based on route information and event information, the optimum route having a higher quality than that designated based on only the route information, and to improve the convenience of nodes.
The above object of the present invention is achieved by a routing system capable of designating an optimum route for a call connection request, the routing system comprises a route information transmitting/receiving unit which exchanges standardized network route information of a local node with adjacent nodes thereof; a route information management unit which manages the standardized network route information as routing information; an event processing unit which manages event information besides the standardized network route information, the event information being obtained by detecting and statistically processing events in the local node; and a routing unit which designates an optimum route for a call connection request based on the routing information managed by the route information management unit and the event information managed by the event processing unit.